Image Quality TransferIQT

Image
Image

Contact Info

+
www.african-cschd.org
UCL - London

IQT Project Team


Dr Matteo Figini

Dr Matteo Figini

Research Associate

Harry Hongxiang Lin

Harry Hongxiang Lin

Research Associate

John

Dr OGBOLE G. I.

Associate Professor



The Organization

The Image Quality Transfer (IQT) project is co-led by an international research group from Nigeria and United Kingdom. In UK, it comprises University College London (UCL) Centre for Medical Image Computing (CMIC) and Department of Computer Science, Great Ormond Street Hospital for Children (GOSH) and UCL Great Ormond Street Institute of Child Health (ICH). Daniel C Alexander leads the overall team, and the other investigators in London are Delmiro Fernandez-Reyes (UCL Department of Computer Science), Helen J Cross and David W Carmichael (UCL GOSH). In Nigeria, team is based at the College of Medicine of the University of Ibadan (COMUI), Nigeria and includes Godwin Ogbole (COMUI Department of Radiology), Biobele J. Brown, and Ikeoluwa Lagunju (COMUI Department of Paediatrics).

The IQT project is funded by “EPSRC Global Challenges Research Fund (GCRF) Diagnostics, Prosthetics and Orthotics panel November 2017”. The funding has been used for software development (learning algorithms, computing environment setup), data acquisition (at UCH Ibadan, at MeCure Healthcare disgnostic center in Lagos, and at GOSH in London), clinical demonstration and evaluation (clinician training and proof-of-concept clinical evaluation).

The long-term vision motivating this project is one of robust, sustainable, and affordable computational solutions to enhance the diagnostic quality of the available imaging resources in low and middle-income countries (LMICs). We provide a proof of concept using MRI from low- field scanners available in LMICs, specifically Nigeria, that we enhance by propagating information from databases of images from state-of-the-art MRI scanners available in the UK. We focus on an application to childhood epilepsy to demonstrate the clinical benefit of IQT. Childhood epilepsy presents an immediate clinical need in LMICs, as MRI from widely available low-field scanners is insufficient to support clinical decisions on curative surgery that are routinely made in the UK using 1.5T or 3T images.

UCL CMIC and Department of Computer Science carries out on developing machine learning algorithms and softwares based on clinical images, public datasets and simulations. GOSH, ICH and COMUI provide MR scans from patients with epilepsy and the radiological and neurological expertise necessary to understand clinical needs.

The Research

We aim at improving the spatial resolution and contrast of low-field MRI images acquired in Nigeria by propagating information from publicly available state-of-the-art 3T MRIs and 1.5T or 3T scans from GOSH. We trained a deep-learning algorithm on paired simulated low-field and real 3T images. In the present phase we successfully applied this algorithm to a small sample of 0.36T data from Nigeria; the preliminary results showed an enhancement of overall image quality and in particular of the visibility of lesions. The developed algorithm will be finally applied to an extensive collection of pediatric epilepsy patients who will undergo MRI scans both at low and high magnetic field. Expert radiologists will compare the enhanced low-field and the high-field images to assess the diagnostic added value of IQT in this field.

We draw on the latest advances in machine learning to approximate the MRIs available in the UK from those accessible in the paediatric neurology clinic in UCH Ibadan, Nigeria - a typical sub-Saharan city hospital. Machine learning has made major advances over the last few years. In particular, it shows remarkable feats of artificial intelligence in data-rich application areas such as computer vision where, for example, computers now outperform humans in object recognition. The advances are just starting to make an impact in medical imaging, which presents unique challenges because a) less data is available than many non-medical computer vision tasks, b) decisions are often more critical as they impact directly on patient outcome.

The project brings together world-leading medical image computing and computer science researchers from UCL with internationally acclaimed experts in pediatric epilepsy from GOSH and COMUI. The University College Hospital Ibadan is a leading research hospital in Nigeria with a proud reputation for innovation, it will be the ideal place to exploit the outcome of this collaboration and transfer the new technology to other LMIC countries where state-of-the-art MRI scanners are not available.

Research Funded by: